jhuneurofandomcom-20200214-history
Cognitive Neuroscience Lecture 8: Object Recognition
Recording info that wasn’t finished last class: ERP, EEG, and MEG are very good temporal resolution but low spatial resolution. Useful when you care about timecourse. Connections beyond V1: we’ve studied what happens in V1 so far There’s more than just patches and bars that we understand. EyeàLGNàv1àsplits to dorsal and ventral path Simplified to two pathways: further processing in dorsal/ventral pathways - dorsal: “where is the object”: spatial - ventral: “what is the object”: identification of object Simplified because there are other things in between; not always dichotomous Monkey experiments are good because they have similar visual systems to us Also good bc can directly test function/role of area by artificially lesioning brain region What vs. where experiment - Task 1: What—'Object discrimination' o Food was hidden under one of them - Task 2: Where—'Landmark discrimination' o Same shape but distance between obj with food and landmark was different than distance from non-food and landmark - Performance good for both tasks in a healthy monkey - Performance bad in landmark task for monkey with dorsal '''lesion - Performance bad in '''object discrimination '''task for monkey with '''ventral lesion - This is double dissociation of object and landmark - Ventral àwhat, dorsalà where/spatial locations is shown by this experiment Double dissociation in Humans - Can’t do artificial lesion, but using neuropsych, studied people with preexisting lesions - Showed selective impairment of tasks - Ventral lesion o CO poisoning o Perception problem: “what” deficit o Visual-form agnosia/object agnosia: '''impaired object recognition NOTECARD o '''Prosopagnosia: '''specific to faces - Dorsal lesion o Stroke o Action problem: “where” deficit o '''Optic Ataxia: '''had trouble reaching to a spatial location NOTECARD - Did similar double dissociation experiment - Task 1: Perception—are these identical? - Task 2: Action—reach out and grab this - Patient with ventral lesion was unable to do Task 1 (perception task), but was able to do Task 2 (action task) - Patient with dorsal lesion was unable to do Task 2 (action task), but unable to do Task 1 (perception task) - Basically, '''dorsal important for action (since orientation is required) and ventral important for perception - Ventral lesion patient: object agnosia—difficulty recognizing objects; but how do we know it’s this? o Alternative hypotheses: It could be a problem with memory of word, difficulty articulating, vision problem, she just can’t name them o We know it’s truly a visual object recognition problem because she was able to tactilely recognize them when they were in her hands o This shows that she knows what they are but just cannot VISUALLY recognize them o Further evidence: § Could draw pictures from memory § Could not recognize or copy pictures given a model picture § Could not 'recognize her own drawings later on o Matching orientation task in ventral lesion pt § '''Explicit matching task '(perceptual orientation matching): Couldn’t match something she was holding to something she was '''looking at § Action task '''(visuomotor “posting”): But when performing the '''action, she could align the object to the slot - Dorsal lesion patient: optic ataxia (opposite of ventral lesion patient) o Fine with recognizing and matching o Difficulty with visually-guided reaching o Make inappropriate grasping movements with hands (does not align hand with slot) - This was a double dissociation between the two patients - DORSAL/ACTION/WHERE: spatial action ''' - '''VENTRAL/PERCEPTION/WHAT - Main point: 'visual processing involves several distinct pathways Further processing in dorsal/ventral pathways '“Where”/action: Dorsal stream NOTECARD - Mostly parietal - V1àSuperior longitudinal fasciculusàposteroparietal cortex ' “What”/perception': Ventral stream NOTECARD - Mostly temporal - V1àInferior longitudinal fasciculusàinferior temporal cortex Oversimplification—connections between them; feeds forward, back to V1, and all directions Despite this, fMRI shows that distinction is somewhat true Evidence for dorsal/ventral pathways NOTECARD 1. Animal lesion studies (monkeys) 2. Human neuropsychological cases (DF&RV) 3. Single unit recordings - What: temporal neurons fire for color, shape, faces o E.g. will fire for shape of hand regardless of orientation, but will fire less when given a less detailed hand shape - Where: parietal neurons '''fire for change in direction of motion, velocity, control of movements of attention o E.g. change if you move your attention to left side to right side o To know that something is moving is to know that it has changed location (where) 4. Human neuroimaging studies with intact subjects - '''Remembering what vs. where o Healthy normal adults o Position task: '“are the objects in the same locations” o '''Object task: '“are these the same objects” o Same stim, diff task: location vs. identity o Object task lights up ventral/temporal area o Position task lights up inferior parietal area o '''Position/object tasks light up different areas even in neurologically intact patients when they were looking at same stimuli o Results of PET study § Dot-location matching used parietal more § Face matching used ventral/temporal more - What/where: color/motion o Color: '''compare brain response of color vs. black&white sq § '''ventral o Motion: 'compare brain response to stationary vs. moving black and white regions § '''Dorsal ' o Results: Color is medial (V4), motion is lateral (MT) '''Lateral occipital complex: object recognition - Lesions in people who cannot do obj recognition are same as location of LOC in intact people - Selectively responds to object information (e.g. not scrambled objects) Further processing in dorsal/ventral Computational problems in object recognition: - Object constancy (car from different views, diff background) ''' o We can recognize objects across changes in illumination, size, occlusion, and viewing position o diff image on retinaàhow do you still recognize? o Does take more time to recognize noncanonical - '''Grouping of features and parts of objects (Gestalt principle) è Help determine what bits of an image belong together o Proximity: closer things get grouped as 1 object o Similarity: similar get grouped o Closure o Good continuation: 2 lines, not scattered o Good form (e.g., regularity, symmetry) - Use of stored object knowledge o We can still identify objects even when there aren’t local cues for grouping o Top-down processing to ID obj § If you already know it’s a dog, you can tell it’s a dog How do we achieve object recognition? Some theories: - Template matching: exemplars ' o Compare input to stored template of an object; yes or no § Then respond whether a match or not § E.g. stored template ‘B’ would match ‘B’ but not ‘A’ o What’s the problem with this? § Doesn’t work for novel objs § You could have something that is still a capital B but is a different font o You can recognize it as something different in a different context o What’s the template for a chair? You recognize many things as 1 type of obj o Also you can categorize things that aren’t evolutionarily coded in a template o '''Problems in template matching ' § Variability of exemplars § New objects § Changing viewpoints § The idea of a grandmother cell rejected · Grandma changes over time, but you can still recognize her; must not have 1 cell for 1 exact face, then. - 'Recognition by components: Geons ' o “Alphabet” of vision § Sphere, block, prisms, etc. § '''Geons: basic building blocks of all object recognition § Objects = Geons + Spatial relationships ' o What geons can’t do: § Some things can’t be reduced to geons easily § Fine with general categories, but not individuals: all would break down into same set of geons, so how do we tell difference between dogs Deficits in object recognition: - Visual agnosia (DF) o Different from memory loss; tactile enables recognition in agnosia, meaning that they ''know ''the object but just can’t recognize it ''visually - Stroke patient: could see visual details and make appropriate hand gestures o Could not describe obj '''functions though; used particular parts to infer it was something else like phone § Could not put parts together into whole o Apperceptive Agnosia ' § Right lateralized (found in people with right damage) § Posteriorà earlier in visual processing stream § Failure in basic perceptual processing · We can tell things from fragmented line drawings, but they cannot o '''Associative Agnosia ' § Left hemisphere § Anterior à Higher level processing failure § Can perceive obj but cannot assign meaning § Fail at matching-by-function · Only can match by form § Category specificity shown by HSV patient · He got common obj 90% correct · but live things only 6% § What are implications of category specific associative agnosia? · We interact more with inanimate things (sensorimotor regions also used, not just vision) o G.S. motor activation to manmade obj o Left premotor, sensory becomes active when viewing tools o Regions: § Extrastriate Body Area (bodies) § Fusiform face area (faces) § Middle Temporal (Motion processing) § Lateral occipital complex (Objects) § Parahippocampal place area (places) § Superior temporal sulcus-face area (faces) · Visual areas are segregated by living vs. inanimate obj o Specificity in ventral visual stream o Large scale segregation in ventral stream for living vs nonliving · Summary of category-specific agnosia: 2 explanations o Motor system engaged when viewing inanimate obj à agnosics ID these better than animate o Ventral visual system is segregated by animate vs inanimate shown in fMRI § Vasculature implies that animate regions more prone to damage in encephalitis/stroke § Animate is usually one lost; we’ve never seen only inanimate lost o 'Integrative Agnosia ' § Right/left hemisphere § Cannot put features into parts, or integrate parts as whole - Category-specific defecits o Apperceptive agnosia o Integrative agnosia – object recognition o Associative agnosia o '''Alexia—written word recognition (reading) deficit o Prosopagnosia—face recognition deficit § Sometimes following damage to FFA § Sometimes congenital § Can recognize that it is a face, but not identity of face § To ID, use non-facial info like hair/clothes o What’s special about faces? § More accurate at identifying whole face than parts · Not true for other obj; just faces § Holistic processing of faces: upside down is hard bc holistic effect disappears § Evidence from object agnosia · Obj recognition difficulties · Normal objects are confusing, but agnosia pts see face · Obj recognition is diff from face recognition § Evidence from single cell recording · There are cells that respond more than others to faces § Fusiform gyrus: Right fusiform, Fusiform Face Area Faces activate it more than scenes''''''